mathoperatorsartificial-intelligencemodal-logic

Can modal operator be defined as a boolean function?


Let us consider for simplicity only Kripke structures with a single agent whose knowledge is described by the modal operator K. We know that in all the corresponding Kripke structures where K is interpreted by equivalence there holds for any formula A

a) the formula KA -> A (Knowledge Axiom) is valid ,
b) but the formulas A -> KA and ¬KA are not valid.

Utilize these facts to show that such a behaviour of the modal operator K cannot be encoded by any boolean function (ie. Truth values defined by a table).

Hint: Suppose the truth value of KA can be calculated from the truth value of A using a truth table for K (in the same way as ¬A is calculated form A). Consider all possible truth tables for K and show that none of them grants the properties a) and b) mentioned above.

I dont understand that hint... making truth table of K is like constructing truth table of negation symbol ¬ , which in my mind doesnt make sense I think it makes sense only to make negation of something and not just negation


Solution

  • Consider all possible truth tables for K:

    | A | K₁A | K₂A  | K₃A | K₄A |
    —————————————————————————————
    | 1 |  1  |  1  |  0  |  0  |
    —————————————————————————————
    | 0 |  1  |  0  |  1  |  0  |
    

    Show that none of them grants the properties a) and b) mentioned above.

    Case 1

    | A | KA  | KA->A | A->KA | ¬KA |
    —————————————————————————————————
    | 1 |  1  |   1   |   1   |  0  |
    —————————————————————————————————
    | 0 |  1  |   0   |   1   |  0  |
    

    In this case, KA->A is not a tautology.

    Case 2

    | A | KA  | KA->A | A->KA | ¬KA |
    —————————————————————————————————
    | 1 |  1  |   1   |   1   |  0  |
    —————————————————————————————————
    | 0 |  0  |   1   |   1   |  1  |
    

    In this case, A->KA is a tautology.

    Case 3

    | A | KA  | KA->A | A->KA | ¬KA |
    —————————————————————————————————
    | 1 |  0  |   1   |   0   |  1  |
    —————————————————————————————————
    | 0 |  1  |   0   |   1   |  0  |
    

    In this case, KA->A is not a tautology.

    Case 4

    | A | KA  | KA->A | A->KA | ¬KA |
    —————————————————————————————————
    | 1 |  0  |   1   |   0   |  1  |
    —————————————————————————————————
    | 0 |  0  |   1   |   1   |  1  |
    

    In this case, ¬KA is a tautology.


    Can desired behaviour of K be encoded by many-valued matrix?

    For alethic modal systems, the answer is the following:

    See, e. g., introductory parts in the article by Jean-Yves Beseau.
    I hope these results are relevant for epistemic modal systems.